Ten years of pioneering research capitalizing on eye-tracking technology to measure spontaneous visual scanning behavior in viewing of naturalistic situations have demonstrated that infants at risk of developing autism spectrum disorders (ASD) display atypical patterns of visual attention in the first year of life, a period when social engagement with faces is crucial for the transition to spoken language. Expanding our programmatic effort in the Emory ACE to track the ontogeny of social engagement and its derailment in autism. Project II will map out the developmental unfolding of social vocal engagement over the first three years of life, to explore the consequences of eariy derailment of social visual engagement investigated in Project I. By tracking the development of spoken language in Project II concurrently with eye-tracking measures of visual attention to talking faces in Project I from 2 to 36 months in a cohort of 235 infants with ASD, non-autistic developmental delays (DD) and typical development (TD), the close synergy between these two projects will shed light on the relationship between two core deficits defining ASD, as well as suggesting mechanisms for targeted eariy intervention exploited in Project III. The goal of Project II is to test the hypothesis that the derailment of emerging communication in infants at risk of autism can be attributed to an eariier breakdown in mechanisms of social engagement. The first aim is to quantify eariy vocal behavior and spoken language development in TD, DD, and ASD infants, based on automated acoustic analysis of day-long audio recordings of each child's language environment collected at monthly intervals from 2-36 months in the home. The second aim is to determine the relationship between social engagement in the first year of life and spoken language development in the first three years of life. The third aim is to determine whether developmental profiles for quantitative measures of visual attention (using eye-tracking measures) and spoken language development (using acoustic measures) within the first two years of life can together be used as: (a) categorical diagnostic markers; (b) predictors of autistic symptomatology; (c) a means of defining endophenotypes within the autism spectrum, linking risk, diagnosis and outcome.